Messages: toolCheckFile & Error in routine 'outDiffKoeff'

Discussion of numerical issues which are not directly linked to specific applications
Post Reply
mauvec
Posts: 19
Joined: Mon Mar 15, 2021 3:38 pm
anti_bot: 333

Messages: toolCheckFile & Error in routine 'outDiffKoeff'

Post by mauvec » Tue Nov 29, 2022 11:31 am

Dear all,

I am utilising the 'profile_from_fil'e feature to describe the temperature evolution in the input file; however the following message is displayed:


File not found: profile-500.txt 1 2
STOP in Routine toolCheckFile


May I inquire what would be the reason for this?

Additionally, the following message is also displayed when the diffusion data is being updated from the database during the first time-step of the simulation,


Error in routine 'outDiffKoeff'


Best regard,

Mauro

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Messages: toolCheckFile & Error in routine 'outDiffKoeff'

Post by Bernd » Tue Nov 29, 2022 12:23 pm

Dear Mauro,

The first error is obvious, you have to use the keyword "columns":

profile-500.txt columns 1 2

The second error message should come with a preceding message line, which could either be:

- "Unable to open file"
- "Error while writing to file"
- "Unable to close file"

Thus it can be anything, from missing disc space or to format overflow or simultaneous access by multiple MICRESS instances. If you can't solve it, please send your input file(s) so that we can try to reproduce the problem.

Bernd

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Messages: toolCheckFile & Error in routine 'outDiffKoeff'

Post by Bernd » Sat Dec 03, 2022 9:28 pm

Dear Mauro,

Thank you for sending your input files as PM. Unfortunately, the .ges5-file is missing, so that I cannot run it. Please send it.

Bernd

mauvec
Posts: 19
Joined: Mon Mar 15, 2021 3:38 pm
anti_bot: 333

Re: Messages: toolCheckFile & Error in routine 'outDiffKoeff'

Post by mauvec » Wed Jan 04, 2023 3:45 pm

Dear Bernd,

I completely forgot sending this. I have pm the .ges5.file.

Bests,

Mauro

mauvec
Posts: 19
Joined: Mon Mar 15, 2021 3:38 pm
anti_bot: 333

Re: Messages: toolCheckFile & Error in routine 'outDiffKoeff'

Post by mauvec » Tue Jan 10, 2023 1:06 pm

Dear Bernd,

May I ask you about any update on this matter? I am still not able to run any simulation due to the same type of error.

Many thanks,

Mauro

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Messages: toolCheckFile & Error in routine 'outDiffKoeff'

Post by Bernd » Wed Jan 11, 2023 10:29 pm

Dear Mauro,

Sorry for the late response. I have run your input file and found some problems with the "profile-500.txt"-file which you provided me by PM: The file contains a lot of trailing Tab-characters which gave me input errors. After cleaning the file, everything is running fine.

Please start cleaning up your file as well - perhaps in your case you have read false temperature values which indirectly created the problem in outDiffKoef...

However, the problem you face could also come from the much too complex diffusion data in the solid phases, leading to format overflow in the outDiffKoeff routine. Keep in mind that diffusion in the solid is practically absent during the time-scale of additive processes and can be omitted or strongly simplified without loss of accuracy. I strongly advise to use diffusion data in a diagonalized way and without concentration dependency, and use segmentation in z-direcion to reduce the extrapolation ranges, as shown in the SLM-examples A017 and A018. Multicomponent diffusion matrixes are very complex in Ni-base-alloys, because fcc is close to a miscibility gap, and even diagonal terms can be negative. This makes extrapolation over large T-ranges extremely risky.

Please try whether you get your simulation running this way, and come back to me if not.

Best wishes
Bernd

Post Reply